Title :
An application of image point operators for OCR improvement in ALPR algorithm
Author :
Lukic, Vojislav ; Makarov, A. ; Spanovic, M.
Author_Institution :
R&D Center, Vlatacom d.o.o. Beograd, Belgrade, Serbia
Abstract :
Optical character recognition (OCR) is a very important part of an automatic license plate recognition (ALPR) algorithm. Often some real scene factors outside the scope of the image acquisition could have significant effect on decrease of the accuracy of the OCR engine. Some of those factors are: filthy and dusty plates, noise and scratch on plate characters, different pixel intensities on different part of the plate image etc. There are various techniques of an image improvement for OCR that are usually computationally very demanding. An ALPR algorithm should be working in a real-time so we try to involve some simple point operators in the license plate image improvement in order to avoid computational costs.
Keywords :
image enhancement; optical character recognition; automatic license plate recognition algorithm; dusty plate; filthy plate; image acquisition; image point operator; license plate image improvement; optical character recognition; pixel intensity; Brightness; Character recognition; Engines; Equations; Image recognition; Licenses; Optical character recognition software; ALPR; OCR; brightnesscontrast; gamma correction; image improvement; point operators;
Conference_Titel :
Telecommunications Forum (TELFOR), 2012 20th
Conference_Location :
Belgrade
Print_ISBN :
978-1-4673-2983-5
DOI :
10.1109/TELFOR.2012.6419305